INDEX

Explanations

The neuron fires on narrative passages describing hands‐on, sensory or classroom activity posts (e.g. “we used the bubble wrap,” “we’re learning about apples,” “after some jumping and popping,” etc.).

writing a report or story

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ּוֹ

-1.30

Psycho

-1.23

Ν

-1.21

ֶּ

-1.19

tomatik

-1.17

 আপ

-1.17

literally

-1.14

ִּ

-1.13

mathbb

-1.13

🤯

-1.12

POSITIVE LOGITS

 vielfäl

1.41

 katholischen

1.39

 τὴν

1.38

 different

1.38

 diferentes

1.34

 velké

1.29

 römischen

1.27

 christlichen

1.26

 using

1.25

 ویتامین

1.25

Activations Density 0.029%